ASSESSMENT OF DESIGN METHODOLOGY AND THREE DIMENSIONAL NUMERICAL (CFD) ANALYSIS OF CENTRIFUGAL BLOWER D. R. Chaudhari 1, H. N. Patel 2 1,2 Mechanical Department, Government Engineering College Dahod, (India) ABSTRACT Case study for assessment of design methodology three dimensional numerical (CFD) analysis of forward curved blade type centrifugal blower. Input data selected for design is based on industrial requirement for texturising machine to study forward curved blade type centrifugal blower. Keywords: Design of Forward Curved Blade Type Centrifugal Blower; Numerical Simulation CFD; Flow Characteristics; I. INTRODUCTION Much research has gone into more more systematic design of centrifugal blower. Different authors have been suggested different procedures, although each has a slightly different method of calculation, the broad underling principles all are similar. Impeller design procedure suggested in this paper is as per Eck Bruno. Casing design Calculation of losses in this design is suggested by author W C Osborne. Volute is taken as spiral shape. Iterations are made to design calculations to get optimum geometry at minimum losses. The design procedure described in three main sections are 1) Non dimensional parameters 2) Impeller design 3) Volute design. The Non dimensional parameters will be of considerable assistance to manufacturers users of fans who are not concerned with the theoretical aspects of designs. Input data is selected common for each design & is based on industrial requirement for texturising machine. The selected input data is, Discharge V =0.5 m 3 /sec, Differential Pressure P=981.2 Pa, Speed N=2800 rpm, Delivery Pressure P d = 784.8 Pa, Suction Pressure P s = - 196.4 Pa, Suction Temperature T s =30 C = 303 K, Atmospheric Pressure=1.01325 x 10 5 Pa, Atmospheric Temperature =30 C = 303 K. The numerical analysis carried out for forward curved blade type centrifugal blower using Ansys CFX software. II. DESIGN OF FORWARD CURVED CENTRIFUGAL BLOWER The design procedure described in three main sections are 1) Non dimensional parameters 2) Impeller design 3) Volute design. The Non dimensional parameters will be of considerable assistance to manufacturers users of fans who are not concerned with the theoretical aspects of designs. 134 P a g e
2.1 Nondimentional Parameters For the design, comparison, critical assessment of all fans, one employs dimensionless coefficients. These coefficients must be dimensionless so that the numerical values which arise are independent of the actual increase in pressure, the mass flow, other physical properties. There are in the fortunate position of being able to recommend a number of dimensionless coefficients which are the results of extensive study which will most probably become stardized. Coordinator has given the relation between in graphical form.the graphical form converted into mathematical form by Weighted Residue Method. The resultant equation for in different stages obtained bellow = a -b where, is diameter coefficient Here, a = 0.99 where 0.1 < < 0.4 = 1.5 where 0.4 < < 2 = 0.995 where 0.1 < < 0.4 b = 0.5866 where 0.4 < < 1 = 0.505 where 1 < < 2 = 2.418 for = 0.443 2.2 Impeller Design Now, = air density, = peripheral velocity at outlet of impeller 0.872 v = volume flow rate, = outlet diameter of impeller 0.159 Formula for the volume coefficient is given in Eck Bruno s book, i..e, where = hub diameter, t = blade thickness = inlet blade angle of impeller. Taking,,, 135 P a g e
Taking values the above calculated, Number of blades in impeller is =outlet blade angle of impeller Thus the number of blades depends only on the the radial ratio. The formula gives an approximate indication of number of blades required for normal radial impellers, while multivane impellers require more specialized treatment. However, the optimum number of blades of radial impeller can only be truly ascertained by experiment. Number of blades optimized experimentally by Mr. Nitin Vibhakar for above taken input data. Thus for In the impeller separation of flow at the bend must be prevented. The most effective measure to combat separation at this point is to accelerate the main stream. Therefore the impeller entry area must be smaller than the intake opening. This change in area will be designed by Thus, where, = blade width at inlet of impeller With 20 % acceleration, i.e. No information is yet available to enable the discharge width to be determined from one definite aspect. In practice, one finds both parallel as well as tapered shrouds enclosing the blades of an impeller. The shape of the shroud will depend on the shape of the blade. The decisive factor is the blade passage, not the meridional area. The mean velocity reduces from to in the blade passage. This deceleration is a very important factor in the design of an impeller. In the absence of reliable experimental data about the separation of flow in rotating passage, one makes use of an analogy with stationary diffusers. Accordingly care must be exercised to ensure that the tapering angle dose not exceed to. Therefore taking the tapering angle, the formula will be where, = blade width at inlet of impeller = blade width at outlet of impeller Now velocity at inlet, where, = peripheral velocity at inlet of impeller = speed of impeller in rpm From the inlet velocity triangle,, Therefore, where, = Relative velocity at inlet of impeller 136 P a g e
Now, where, = absolute velocity at inlet of impeller One plots the cross- sectional area of the shroud or diameter of an equivalent circle against the mean velocity of the stream lines so that the enlargement of the blade passage area can be easily examined. The shorter the blade passages, i.e. a larger value for, the less area. In general, one can relate the permissible enlargement of the blade passages, which is dependent on, to the diameter ratio. Therefore as an approximation So assumed that where, = Relative velocity at inlet of impeller From outlet velocity diagram, where, = Tangential component of Absolute Velocity Now, where, = absolute velocity at outlet of impeller Air angle at outlet is, The Fan Power = P V 981.2 0.5 490.6Watts Consider power factor 1.1 Therefore, Power factor P = m x W 1.1 thefan power =1.1x 490.6 =539.66 Watts 2.3 Design of Volute Casing Analyzing steady flow energy equation at inlet exit 1. Casing Outlet Velocity 2. Width of volute casing (Bv), Backward curved is 2.5 b 2, Forward curved is 1.25 b 2 And For Radial tipped is taken as 1.875 b 2 = 2 b 2 Here Bv = 1.8 b 2 = 2 x 0.0433 = 0.055 m 137 P a g e
3. Centrifugal Fan Casing Where k = 0.0023 for backward blade impeller = 0.0020 for forward blade impeller = 0.00215 for radial blade impeller Therefore for forward blade impeller So we get, in º 0 60 120 180 240 300 360 r in m 0.1515 0.16968 0.18786 0.20604 0.22422 0.2424 0.26058 4. Radius of volute tongue 5. Angle of volute tongue 2.4 Losses 1. Leakage loss ( ), Here, We are dealing with sharp-edged openings we shall assume that a coefficient of contraction is applicable to gap. Thus volume leakage through the gap is 2. Entry losses ( ), this type of Losses may arise from a change of direction in the impeller, i.e. upon entry into the impeller intake; the air is diverted through an angle of approximately 90 before entry into cascade. These losses, which are comparable to losses at bends, are dependent upon the values of. If one relates this loss in the usual manner to the dynamic pressure of the maximum velocity, Then Where, 3. Friction loss in the impeller ( ), the greatest losses arise from the passage of a fluid through an impeller. Length of blade curve is Where, Where, 138 P a g e
4. Volute Casing Pressure Loss( ) 5. Disk Friction( ) Other Parameters: 1. Power Loss InWatts Due To Disk Friction 2. Hydraulic Efficiency 3. Volumetric Efficiency 4. Total efficiency 5. Power Required to Run the impeller So Torque 6. Shaft Diameter 7. Blade Profile by Circular arc Method Table 2.1 shows the values as per 0-iteration. Now, further calculations are done by adding leakage loss pressure losses in discharge differential pressure given in design input data respectively. 1 st 2 nd iterations are shown in table 2.2, table 2.3 respectively. The calculation summaries listed bellow in tables. 139 P a g e
0 th ITERATION At Table 2.1 0 th ITERATION At Non dimensional parameters Volute Casing Speed coefficient 0.443 Width Of Casing 0.055 m diameter coefficient 2.418 Outlet Velocity Of Casing 21.87 m/s Pressure coefficient 0.872 Diameter Of Casing at 0 0.1515 m Volume coefficient 0.159 Diameter Of Casing at360 0.2605 m Impeller inlet Dimensions Volute Tongue Angle 7 Peripheral Velocity 30.49 m/s Radius of Tongue 0.1623 m Relative Velocity 35.212 m/s Casing Pressure Losses 10 pa Meridian Velocity 17.606 m/s Disk Friction 0.0066 N m Absolute Velocity 17.606 m/s Power Loss Disk Friction 1.995 watt Impeller Diameter 0.208 m Power Required To Run impeller 584.8watt Width Of Blade 0.0433 Hydraulic Efficiency 89.45 Air Angle 90 Volumetric Efficiency 81.61 Blade Angle 30 Total Efficiency 73.8 Impeller Outlet Dimensions Shaft Diameter 0.01069 m Peripheral Velocity 43.95m/s Blade Profile Radius 0.1894 m Relative Velocity Meridian Velocity Absolute Velocity Impeller Diameter Width Of Blade 24.252 m/s 14.385 m/s 28.323 m/s 0.302 m 0.035 m Air Angle 30.46 Blade Angle 36.30 Leakage Loss 0.113 Entry loss Pressure Losses In Impeller 27.08 pa 78.60 pa. 140 P a g e
1 st ITERATION At Table 2.2 1 st ITERATION At Non dimensional parameters Volute Casing Speed coefficient 0.4508 Width Of Casing 0.059 m diameter coefficient 2.3937 Outlet Velocity Of Casing 22.5 m/s Pressure coefficient 0.860 Diameter Of Casing at 0 0.1605 m Volume coefficient 0.162 Diameter Of Casing at360 0.2760 m Impeller inlet Dimensions Volute Tongue Angle 7 Peripheral Velocity 32.7 m/s Radius of Tongue 0.172 m Relative Velocity 37.76 m/s Casing Pressure Losses 15.52 pa Meridian Velocity 17.86 m/s Disk Friction 0.0072 N m Absolute Velocity 17.86 m/s Power Loss Disk Friction 2.11 watt Impeller Diameter 0.2225 Power Required To Run impeller 803.6watt Width Of Blade 0.0465 Hydraulic Efficiency 89.3 Air Angle 90 Volumetric Efficiency 82.71 Blade Angle 30 Total Efficiency 73.86 Impeller Outlet Dimensions Shaft Diameter 0.01170 m Peripheral Velocity 46.79 m/s Blade Profile Radius 0.1933 m Relative Velocity Meridian Velocity Absolute Velocity Impeller Diameter Width Of Blade 26.34 m/s 15.32 m/s 29.80 m/s 0.320 m 0.038 m Air Angle 30.99 Blade Angle 35.37 Leakage Loss 0.128 Entry loss Pressure Losses In Impeller 28.31 pa 87.03 pa 141 P a g e
2 nd ITERATION At Table 2.3 2 nd ITERATION At Non dimensional parameters Volute Casing Speed coefficient 0.4465 Width Of Casing 0.059 m diameter coefficient 2.4071 Outlet Velocity Of Casing 22.5 m/s Pressure coefficient 0.866 Diameter Of Casing at 0 0.1610 m Volume coefficient 0.161 Diameter Of Casing at360 0.2769 m Impeller inlet Dimensions Volute Tongue Angle 7 Peripheral Velocity 32.75 m/s Radius of Tongue 0.1723 m Relative Velocity 37.87 m/s Casing Pressure Losses 15.52 pa Meridian Velocity 17.86 m/s Disk Friction 0.00725 N m Absolute Velocity 17.86 m/s Power Loss Disk Friction 2.112 watt Impeller Diameter 0.223 Power Required To Run impeller 803.6watt Width Of Blade 0.0465 Hydraulic Efficiency 89.4 Air Angle 90 Volumetric Efficiency 82.71 Blade Angle 30 Total Efficiency 73.86 Impeller Outlet Dimensions Shaft Diameter 0.01170 m Peripheral Velocity 47.01 m/s Blade Profile Radius 0.1964m Relative Velocity Meridian Velocity Absolute Velocity Impeller Diameter Width Of Blade 26.34 m/s 15.355 m/s 29.86 m/s 0.321 m 0.0385 m Air Angle 30.94 Blade Angle 35.66 Leakage Loss 0.1286 Entry loss Pressure Losses In Impeller 28.31 pa 87.84 pa After this diameter of impeller & the parameter is unchanged. So, there is no need to do further iteration. 142 P a g e
III. NUMERICAL ANALYSIS RESULTS The numerical analysis carried out for forward curved blade type centrifugal blower using Ansys CFX software is presented herewith. This section presents qualitative quantitative simulation results of the flow in forward curved blade type centrifugal blower with 0.5 m3/s discharge, 2800 rpm rotational speed 16 numbers of blades in impeller. Efficient energy transfer in a centrifugal blower depends upon proper blade profile, gradual change in area of volute casing smooth surface finish. For such energy transfer Flow lines must be parallel to each other should generate streamlined flow within guided three dimensional passages. Smooth parallel streamlines within around impeller region confirms well guided path for flow offered by this design. Flow leaves impeller smoothly to enter in volute casing. Volute casing progressively transmit flow up to outlet of the centrifugal blower. The flow just before tongue is recirculating towards impeller after tongue it generates small vortex. Following figure shows the streamline pattern for forward curved centrifugal blower with 0.5 m3/s discharge, 2800 rpm rotational speed 16 number of blades in impeller Figure: Streamlines of Designed Centrifugal Blower Contours of static pressure, total pressure velocity magnitude at inlet of inlet duct of forward curved blade centrifugal blower Range for static pressure = -206.4 to -198.7 Pa Average static pressure = -203.102 pa Range for total pressure = -178.4 to -170.8 Pa Average total pressure = -175.039 pa Range for velocity magnitude = 0 to 6.884 m/s Average velocity magnitude = 6.88213 m/s 143 P a g e
Contours of static pressure, total pressure velocity magnitude at inlet of impeller of forward curved blade centrifugal blower Range for static pressure = -2796 to 69.25 Pa Average static pressure = -482.403 pa Range for total pressure = -3347 to 290.6Pa Average total pressure = -491.022 pa Range for velocity magnitude = 0 to 52.46 m/s Average velocity magnitude = 22.8954 m/s Contours of static pressure, total pressure velocity magnitude at outlet of impeller of forward curved blade centrifugal blower Range for static pressure = 457.2 to -1341 Pa Average static pressure = -613.294 pa Range for total pressure = -2495 to -56.65Pa Average total pressure = -1421.74 pa Range for velocity magnitude = 0 to 44.46 m/s Average velocity magnitude = 27.6072 m/s Contours of static pressure, total pressure velocity magnitude at outlet of casing of forward curved blade centrifugal blower Range for static pressure = 457.2 to -1341 Pa Average static pressure = -613.294 pa Range for static pressure = 457.2 to -1341 Pa Average static pressure = -613.294 pa Range for static pressure = 457.2 to -1341 Pa Average static pressure = -613.294 pa 144 P a g e
Taken input data for design of blower, i.e. Discharge = 0.5 m3/s, Differential pressure =981.2, Rotational speed = 2800 rpm, Inlet pressure at inlet duct = -196.4 pa Outlet pressure at casing = 784.8 pa And simulation of flow in forward curved blade type blower at rotational speed = 2800 rpm at discharge =0.5 m3/s gives following results, i.e. Inlet pressure at inletduct = -203.102 pa, Outlet pressure at casing = 785.085 pa, Differential pressure = 988.187 pa So we can say error between simulated results in design of blower is = 0.707% IV. CONCLUSION 1. The theoretical numerical analysis (CFD) is closer to design point conditions in centrifugal blower under study. 2. The flow phenomenon of recirculation near tongue region is confirmed by numerical analysis as shown by stream line diagram. 3. Pressure pulsations are observed at impeller outlet near tongue region caused by obstruction of tongue. Hence design of tongue is very important in blower design to reduce back flow recirculation. 4. The mean pressure distribution around the volute is not uniform even at the design flow rate. Jet wakes are observed in the vicinity of tongue region. 5. The nature of curves obtained after simulation closely follows trend of stard fan performance curves 6. Low high pressure regions along suction pressure side respectively of a blade are visualized by numerical analysis. Energy transfer from impeller to fluid is also confirmed by pressure velocity contours within blade passage. REFERENCES Books: [1] Bruno. Ek. 1975, Fans Design & Operation of Centrifugal, Axial Flow Cross Flow Fans 1 st Edition, Pergamon Press. [2] Osborne.W.C, 1966, Fans, Pergamon Press Theses: [3] N. N. Vibhakar. Experimental Investigations On Radial Tipped Centrifugal Blower. M.E. Dissertation Report Submitted To South Gujarat University, 1998. [4] D. M. Sharma. Experimental And Analytical Investigations Of Slip Factor In Radial Tipped Centrifugal Blower. M.E. Dissertation Report Submitted To South Gujarat University, 2000. [5] C. C. Patel. Comparative Assessment Of Design Methodologies For Radial Tipped Centrifugal Blower. M.E. Dissertation Report Submitted To South Gujarat University, 2001. 145 P a g e
[6] K. Shah. Unified Design And Comparative Performance Evaluation Of Forward And Backward Curved Radial Tipped Centrifugal Blower/Fan. M.E. Dissertation Report Submitted To South Gujarat University, 2002. [7] A. Dube. Development Of Software In VB For Unified Design Method Of Radial Tipped Centrifugal Blowers. M.E. Dissertation Report Submitted To South Gujarat University, 2003. [8] T. Balaganeshan. Experimental Investigations Towards Establishing A Loss Model For Radial Tipped Centrifugal Blower/Fan. M.E. Dissertation Report Submitted To South Gujarat University, 2003. [9] Suhas Dileep Masutage Three Dimensional Numerical (CFD) Analysis of Backward & Forward Curved Radial Tipped Blade Centrifugal Fan with Varying Number of Blades Journal Papers: [10] Sheam-Chyun Lin, Chia-Lieh Huang, An integrated experimental numerical study of forward curved centrifugal fan [11] Martin Gabi Toni Klemm, Numerical experimental investigations of cross-flow fans 146 P a g e